Biologic rationale, biomarker precision, and efficacy results in the biomarker-negative population all helped FDA officials decide the final indication.
Data from the ongoing CAPItello-291 trial led to the November 2023 approval of capivasertib (Truqap; AstraZeneca) in combination with fulvestrant (Faslodex; AstraZeneca) for the treatment of hormone receptor (HR)–positive, HER2-negative locally advanced or metastatic breast cancer with 1 or more PIK3CA/AKT1/PTEN alterations, according to a session at the 2023 San Antonio Breast Cancer Symposium.
Presenter Christy Osgood, MD, a medical officer with the FDA, reviewed the trial design, findings, and regulatory decision-making that went into this approval. The 708 trial participants were diagnosed with HR-positive, HER2-negative unresectable or metastatic breast cancer and had received up to 2 prior lines of endocrine therapy and up to 1 prior line of chemotherapy. Patients were stratified by liver metastases, prior cyclin-dependent kinase 4/6 (CDK4/6) inhibitors, and geographical region.
The trial had dual primary end points of PFS in the overall population as well as the PIK3CA/AKT1/PTEN biomarker-positive population. Secondary endpoints included overall survival (OS) in the overall population as well as the PIK3CA/AKT1/PTEN biomarker-positive population, overall response rate, and duration of response.
When considering indications, Osgood said the FDA must evaluate benefit in both the biomarker-positive and -negative populations. Officials have been encouraging sponsors to design trials for targeted agents so that they are adequately powered to detect the primary end point in both populations, but in the absence of this she said the FDA will perform exploratory subgroup analyses to determine the favored population.
In the CAPItello-291 population, Osgood said several key factors helped determine that the treatment favored the biomarker-positive population. These included biologic rationale, biomarker precision, efficacy results in the biomarker-negative population, differences in efficacy between the biomarker-positive and -negative groups, the safety profile, and availability and consistency of external data.
“Even if the analysis in the biomarker-negative population is an exploratory analysis, if there is minimal efficacy in the biomarker-negative subgroup and relatively more efficacy in the biomarker-positive subgroup, then it may be determined that the benefit in the overall population is primarily driven by the biomarker-positive population, and this would favor granting an indication in the biomarker-positive population only,” Osgood explained.
In the trial, 44 patients in the investigational arm were biomarker-positive, compared to 38 patients in the placebo-plus-fulvestrant arm. Additionally, 40 patients in the investigational arm were biomarker-negative compared to 48 in the control arm.
The trial found statistically significant improvements in PFS in the overall population, with a median PFS of 7.2 months in the capivasertib arm versus 3.6 months in the control arm. Improvement was also demonstrated in the biomarker-positive population, with a median PFS of 7.3 months with capivasertib versus 3.1 months with placebo.
Notably, Osgood pointed out that median PFS between the 2 arms among those who were biomarker-negative differed by only approximately 5 days, highlighting the potential lack of clinical meaningfulness.
Interim OS analysis showed a 0.91 hazard ratio in the biomarker-negative population. Notably, Osgood said Kaplan-Meier curves in the biomarker-negative population crossed multiple times, making any conclusions unreliable.
Adverse events (AEs) of special interest were hyperglycemia, diarrhea, and cutaneous AEs. Hyperglycemia of all grades occurred in 18% of the treatment arm compared to 4.3% of the placebo arm; diarrhea of any grade occurred in 72% and 20%, respectively; and cutaneous AEs of any grade occurred in 58% and 18%, respectively. Osgood also noted that there were more grade ³3 AEs in the capevasertib arm.
In addition to these findings from the CAPItello-291 trial, Osgood said external data from the FAKTION study further supported the final indication. The FAKTION study found potentially less activity in the biomarker-negative patient population, making the CAPItello-291 findings less likely to be due to chance alone.
“Based on all of this information, the risk-benefit is not thought to be favorable in the overall population and the efficacy appeared to be more narrowly attributed to the biomarker-positive [population],” Osgood concluded.
Osgood C. Special Session 1: New Drug Approvals for Metastatic Breast Cancer. Presented at: San Antonio Breast Cancer Symposium. December 5-9, 2023.